My starting point, at last!!!

The first argument I will treat is an important step particularly looking at in vitro toxicology. The question is: it is better one experiment with a large number of replicates or a large number of experiments with few replicates? And how can I take into account both of them? This point is only marginally treated in literature and the statistical procedures poorly explained in almost all scientific studies. The number of experiments and replicates are almost always reported, but statistical analysis is only summarily described as dependent/independent t-student or ANOVA depending on the number of conditions to be compared. The problem is more complex than it appears. Despite a good repeatability of an experiment is mandatory, and if it is not ensured the used assay/technique should be carefully verified, so that possible confounding factors can merge, experiments can vary both on the crude values and on the variability of replicates (Standard Deviation of single experiments). I have already explained some aspects of this problem in Goldoni M, Tagliaferri S. Dose-response or dose-effect curves in in vitro experiments and their use to study combined effects of neurotoxicants. Methods Mol Biol. 2011;758:415-34. However, this book chapter is more focused on other aspects of toxicology and it is not freely available. In the next weeks, I will try to introduce some aspects of the available methods, using also practical examples to better explain the concepts and introducing some commands of SPSS or other statistical programs. Stay tuned!